CAPE: A much stronger predictor of stock returns than many think

Discuss all general (i.e. non-personal) investing questions and issues, investing news, and theory.
User avatar
Topic Author
willthrill81
Posts: 32250
Joined: Thu Jan 26, 2017 2:17 pm
Location: USA
Contact:

Re: CAPE: A much stronger predictor of stock returns than many think

Post by willthrill81 »

Forester wrote: Sat Apr 10, 2021 5:36 am US CAPE is now over 37, highest since the year 2000 https://www.multpl.com/shiller-pe/table/by-month

The real 10yr yield today is -0.64%, back then it was 1.77%.
The latter is a likely contributor to the former.
The Sensible Steward
foodhype
Posts: 190
Joined: Thu May 25, 2017 10:38 pm

Re: CAPE: A much stronger predictor of stock returns than many think

Post by foodhype »

For issues with CAPE and possible solutions see viewtopic.php?f=10&t=345664
Random Walker
Posts: 5561
Joined: Fri Feb 23, 2007 7:21 pm

Re: CAPE: A much stronger predictor of stock returns than many think

Post by Random Walker »

willthrill81 wrote: Sat Apr 10, 2021 9:38 am
Forester wrote: Sat Apr 10, 2021 5:36 am US CAPE is now over 37, highest since the year 2000 https://www.multpl.com/shiller-pe/table/by-month

The real 10yr yield today is -0.64%, back then it was 1.77%.
The latter is a likely contributor to the former.
Great website; SOBERING STATISTIC!

Dave
Random Walker
Posts: 5561
Joined: Fri Feb 23, 2007 7:21 pm

Re: CAPE: A much stronger predictor of stock returns than many think

Post by Random Walker »

am wrote: Sat Apr 10, 2021 8:56 am
Forester wrote: Sat Apr 10, 2021 5:36 am US CAPE is now over 37, highest since the year 2000 https://www.multpl.com/shiller-pe/table/by-month

The real 10yr yield today is -0.64%, back then it was 1.77%.
What to do? Probably nothing. How about if cape is in record territory or in the 50s or 60s? Does low bond yields invalidate cape?

As I approach my goal for investments years before I’m ready to retire, I wonder whether I should dial back risk from about 80/20 to 70/30 or less? On the other hand , I’ve hopefully got decades of investing left.
A Monte Carlo Simulation could be very useful for you in making the decisions. I’m guessing you might be surprised by the lack of sensitivity the likelihood of meeting your goals is to asset allocation. You may actually find less aggressive allocations are more likely to achieve your goals while resulting in lower terminal value. More aggressive allocation will result in higher average terminal value, but significantly wider dispersion of potential outcomes.

Dave
User avatar
HomerJ
Posts: 21281
Joined: Fri Jun 06, 2008 12:50 pm

Re: CAPE: A much stronger predictor of stock returns than many think

Post by HomerJ »

Forester wrote: Sat Apr 10, 2021 5:36 am US CAPE is now over 37, highest since the year 2000 https://www.multpl.com/shiller-pe/table/by-month

The real 10yr yield today is -0.64%, back then it was 1.77%.
In 1996, Forester, you have would written.

US CAPE is now over 25, highest since 1929!!

The real 10yr yield today is 1.8%, back then it was 5%.

--------------------

Yet 1996 was an good time to buy. Not just an average time. At the highest valuations in history since the Great Depression, 1996 turned out to be an GOOD time to buy. How is that possible if CAPE is so predictive?

S&P 500 was around 650 through most of 1996. It more than doubled to 1485 by 2000. During the dot-com crash, it dropped to 837.

So 1996, the highest valuations in history since the Great Depression turned out be a CHEAP time to buy stocks. They have never been cheaper.

Is today 1996? or 2000?

No one knows. Wrap your head around the fact that you don't know.

I'm not saying you're wrong that a crash may be coming soon; I'm not taking the opposite position from you. I'm saying you (and I) don't know.
"The best tools available to us are shovels, not scalpels. Don't get carried away." - vanBogle59
Robot Monster
Posts: 4215
Joined: Sun May 05, 2019 11:23 am

Re: CAPE: A much stronger predictor of stock returns than many think

Post by Robot Monster »

Random Walker wrote: Sat Apr 10, 2021 10:04 am
willthrill81 wrote: Sat Apr 10, 2021 9:38 am
Forester wrote: Sat Apr 10, 2021 5:36 am US CAPE is now over 37, highest since the year 2000 https://www.multpl.com/shiller-pe/table/by-month

The real 10yr yield today is -0.64%, back then it was 1.77%.
The latter is a likely contributor to the former.
Great website; SOBERING STATISTIC!

Dave
Why not look at the Earnings CAPE yield (ECY), also developed by Shiller, which is a lot less scary--or not scary at all, maybe. link
am
Posts: 4233
Joined: Sun Sep 30, 2007 9:55 am

Re: CAPE: A much stronger predictor of stock returns than many think

Post by am »

Random Walker wrote: Sat Apr 10, 2021 10:11 am
am wrote: Sat Apr 10, 2021 8:56 am
Forester wrote: Sat Apr 10, 2021 5:36 am US CAPE is now over 37, highest since the year 2000 https://www.multpl.com/shiller-pe/table/by-month

The real 10yr yield today is -0.64%, back then it was 1.77%.
What to do? Probably nothing. How about if cape is in record territory or in the 50s or 60s? Does low bond yields invalidate cape?

As I approach my goal for investments years before I’m ready to retire, I wonder whether I should dial back risk from about 80/20 to 70/30 or less? On the other hand , I’ve hopefully got decades of investing left.
A Monte Carlo Simulation could be very useful for you in making the decisions. I’m guessing you might be surprised by the lack of sensitivity the likelihood of meeting your goals is to asset allocation. You may actually find less aggressive allocations are more likely to achieve your goals while resulting in lower terminal value. More aggressive allocation will result in higher average terminal value, but significantly wider dispersion of potential outcomes.

Dave
That’s a good point. Terminal value of portfolio is much less important to me then meeting my portfolio goal at retirement and then having lower likelihood of running out of money. Sounds like lower equities may be better? Although some studies show higher equity allocation is better at not running out of money?
YRT70
Posts: 1289
Joined: Sat Apr 27, 2019 8:51 am

Re: CAPE: A much stronger predictor of stock returns than many think

Post by YRT70 »

Perhaps interesting to some here, Ben Felix recently commented on the predictive power of CAPE on his forum.
Basically, the predictive power of CAPE goes away if you consider independent samples (non overlapping) instead of rolling periods. It wasn’t this paper that changed my mind (it was a long talk with someone from Dimensional’s research group) but this paper touches the same points.

https://papers.ssrn.com/sol3/papers.cfm ... id=3142575

[...]

My current understanding is that we should treat expected returns as if CAPE has no information because of how weak the evidence of predictability is. We had previously been devising a fancy expected returns model which included CAPE as a predictor but after the aforementioned conversation I think that we will revert back to bootstrapping from the DMS data back to 1900 to estimate expected returns.
HootingSloth
Posts: 1050
Joined: Mon Jan 28, 2019 2:38 pm

Re: CAPE: A much stronger predictor of stock returns than many think

Post by HootingSloth »

Random Walker wrote: Sat Apr 10, 2021 10:11 am A Monte Carlo Simulation could be very useful for you in making the decisions. I’m guessing you might be surprised by the lack of sensitivity the likelihood of meeting your goals is to asset allocation. You may actually find less aggressive allocations are more likely to achieve your goals while resulting in lower terminal value. More aggressive allocation will result in higher average terminal value, but significantly wider dispersion of potential outcomes.

Dave
Dave,

I realized that I had never really done this myself despite spending a lot of time playing around with Portfolio Visualizer. This was an interesting exercise for me as well, and I'm glad you suggested it.

I was somewhat surprised how little of a difference asset allocation made in our circumstances, even switching between 80/20 and 20/80 over the span of multiple decades. Obviously, the "good" tail was pretty different, but a lot of the results were more in the "Would I really notice this difference?" range than I expected. 80/20 went from "That seems like enough" to "Maybe I should start reading about estate planning and NetJets" while 20/80 went from "I won't worry about whether this is enough" to "This will be a very luxurious lifestyle." And that's even with the aggressively fat tails that you get from Monte Carlo simulations.

It was helpful in reaffirming that a lot of these decisions are less important than just avoiding some big behavioral mistake.
Global Market Portfolio + modest tilt towards volatility (80/20->60/40 as approach FI) + modest tilt away from exchange rate risk (80% global+20% U.S. stocks; currency-hedge bonds) + tax optimization
Random Walker
Posts: 5561
Joined: Fri Feb 23, 2007 7:21 pm

Re: CAPE: A much stronger predictor of stock returns than many think

Post by Random Walker »

HootingSloth wrote: Sat Apr 10, 2021 9:03 pm
Random Walker wrote: Sat Apr 10, 2021 10:11 am A Monte Carlo Simulation could be very useful for you in making the decisions. I’m guessing you might be surprised by the lack of sensitivity the likelihood of meeting your goals is to asset allocation. You may actually find less aggressive allocations are more likely to achieve your goals while resulting in lower terminal value. More aggressive allocation will result in higher average terminal value, but significantly wider dispersion of potential outcomes.

Dave
Dave,

I realized that I had never really done this myself despite spending a lot of time playing around with Portfolio Visualizer. This was an interesting exercise for me as well, and I'm glad you suggested it.

I was somewhat surprised how little of a difference asset allocation made in our circumstances, even switching between 80/20 and 20/80 over the span of multiple decades. Obviously, the "good" tail was pretty different, but a lot of the results were more in the "Would I really notice this difference?" range than I expected. 80/20 went from "That seems like enough" to "Maybe I should start reading about estate planning and NetJets" while 20/80 went from "I won't worry about whether this is enough" to "This will be a very luxurious lifestyle." And that's even with the aggressively fat tails that you get from Monte Carlo simulations.

It was helpful in reaffirming that a lot of these decisions are less important than just avoiding some big behavioral mistake.
I’ve found MCS to be enlightening. Like I said, I was surprised to see the lack of sensitivity of results to the overall asset allocation. Implicit in the exercise, it really focuses the investor on clarifying needs versus wants. Ive found that focus valuable. I guess in the financial advisor business they divide goals into needs, wants, and wishes.

Avoiding big behavioral mistakes is huge. I ultimately evolved from a DIY TSM Boglehead to a Factorhead using an advisor. I did a lot of spreadsheets in making the decision, and I even tried to factor in avoidance of behavioral errors. May have helped me arrive at the result I was looking for, but I thought avoidance of behavioral errors dominated expense ratio differences. A couple or few behavioral errors
in an investing career can make the expense ratio differences we argue about here look silly.

A lot of people here talk about cooling off their portfolio from 80/20 to 70/30 or 60/40 in retirement. When even a 60/40 TSM portfolio has more than 85% of its risk wrapped up in the market factor, I don’t think they are cooling off the portfolio as much as they think. It’s worthwhile to use MCS to really look at how much or how little bigger differences in the basic AA, as you did from 20/80 to 80/20, affect outcomes. The trade off for pretty well off individuals between mean terminal value and likelihood of not outliving money is eye opening.

Dave
bog007
Posts: 539
Joined: Fri Feb 28, 2020 2:27 am

Re: CAPE: A much stronger predictor of stock returns than many think

Post by bog007 »

Alchemist wrote: Sat Apr 10, 2021 7:20 am Cliff Asness is kind enough to provide us an easily tested real world application. In November of 2012, he used CAPE to predict the return of the S&P 500 over the following decade. The prediction still has a year an half or so to reach a full decade but we can examine how it is going so far.

Asness calculated CAPE prediction: 0.9%

Actual return of VFIAX up to March 2021: 15.36%

The prediction was off by 15.25% :oops:

To be fair to Cliff, his prediction ranged from -4.4% to +8.3% so it is a ‘mere’ 7.06% outside of the already pretty useless 12.7% error range. There also could be some horrific market crash between today and November 2022. It will be interesting to come back and check on how this prediction of his fully fleshes out.

Link for Asness prediction: https://www.aqr.com/Insights/Research/W ... Shiller-PE

Portfolio Visualizer source for VFIAX: https://www.portfoliovisualizer.com/bac ... ion1_1=100
and 2 months after that article the market goes up 35% in 2013
Don’t let anyone else ruin your portfolio. It’s your portfolio. Ruin it yourself!!!
User avatar
HomerJ
Posts: 21281
Joined: Fri Jun 06, 2008 12:50 pm

Re: CAPE: A much stronger predictor of stock returns than many think

Post by HomerJ »

Alchemist wrote: Sat Apr 10, 2021 7:20 am Cliff Asness is kind enough to provide us an easily tested real world application. In November of 2012, he used CAPE to predict the return of the S&P 500 over the following decade. The prediction still has a year an half or so to reach a full decade but we can examine how it is going so far.
This is why it drives me nuts when people bring out charts trying to prove that CAPE is predictive.

With new data, we get new charts that follow new straight lines, and people point to that as proof.

But those of us who have been around for the past 30 years have seen and read the ACTUAL predictions made in the past (based on the data they had at the time), and almost every single prediction has been terrible.

Not a little bit off... but WAY off.

Like Cliff... he said 0.9% 10-year returns.

Hedged his bets and said -4% to +8%.

Instead we've gotten 15%

I mean, if we had gotten 7%-8%, maybe his model was still correct and we could have just gotten lucky, sure... But 15% when you predict a mean of 0.9%??!!

AND STILL we have people claiming valuations are predictive and useful?

Absolutely drives me nuts.

Don't think that I am saying there won't be a crash. I do believe in cycles. Bear markets followed by bull markets followed by bear markets followed by bull markets. It does make sense that high valuations seem to indicate we are near the "top"

But the definition of "high valuations" continually changes. In 1996, a CAPE of 25 was EXTREMEMLY high. Now, it's normal.

I don't believe in back-tested models, especially ones with only one or two variables. Economics is far more complex than that.

Valuations are not actionable. Don't make changes based on valuations. Accept there will be a crash, maybe tomorrow, maybe in 5 years, but you better be prepared for it to happen tomorrow. And this is ALWAYS true. Even if valuations were "normal" or even "low", we could still have a crash, or a long period of poor returns.

Just like we've had a long period of good returns even with valuations supposedly being "high" the entire time.

I always plan around the possibility of low returns for the next 10 years, and then I no longer have to care about valuations or any stupid predictions made by PhDs who think they can predict the future.

(They probably think this because they NEVER get called out on it - Cliff Asness will never be called to account for his bad 2012 prediction or his funds poor performance - instead he will always be introduced as an investing genius everywhere he goes)
"The best tools available to us are shovels, not scalpels. Don't get carried away." - vanBogle59
Nathan Drake
Posts: 6234
Joined: Mon Apr 11, 2011 12:28 am

Re: CAPE: A much stronger predictor of stock returns than many think

Post by Nathan Drake »

HomerJ wrote: Sun Apr 11, 2021 11:50 am
Alchemist wrote: Sat Apr 10, 2021 7:20 am Cliff Asness is kind enough to provide us an easily tested real world application. In November of 2012, he used CAPE to predict the return of the S&P 500 over the following decade. The prediction still has a year an half or so to reach a full decade but we can examine how it is going so far.
This is why it drives me nuts when people bring out charts trying to prove that CAPE is predictive.

With new data, we get new charts that follow new straight lines, and people point to that as proof.

But those of us who have been around for the past 30 years have seen and read the ACTUAL predictions made in the past (based on the data they had at the time), and almost every single prediction has been terrible.

Not a little bit off... but WAY off.

Like Cliff... he said 0.9% 10-year returns.

Hedged his bets and said -4% to +8%.

Instead we've gotten 15%

I mean, if we had gotten 7%-8%, maybe his model was still correct and we could have just gotten lucky, sure... But 15% when you predict a mean of 0.9%??!!

AND STILL we have people claiming valuations are predictive and useful?

Absolutely drives me nuts.

Don't think that I am saying there won't be a crash. I do believe in cycles. Bear markets followed by bull markets followed by bear markets followed by bull markets. It does make sense that high valuations seem to indicate we are near the "top"

But the definition of "high valuations" continually changes. In 1996, a CAPE of 25 was EXTREMEMLY high. Now, it's normal.

I don't believe in back-tested models, especially ones with only one or two variables. Economics is far more complex than that.

Valuations are not actionable. Don't make changes based on valuations. Accept there will be a crash, maybe tomorrow, maybe in 5 years, but you better be prepared for it to happen tomorrow. And this is ALWAYS true. Even if valuations were "normal" or even "low", we could still have a crash, or a long period of poor returns.

Just like we've had a long period of good returns even with valuations supposedly being "high" the entire time.

I always plan around the possibility of low returns for the next 10 years, and then I no longer have to care about valuations or any stupid predictions made by PhDs who think they can predict the future.

(They probably think this because they NEVER get called out on it - Cliff Asness will never be called to account for his bad 2012 prediction or his funds poor performance - instead he will always be introduced as an investing genius everywhere he goes)

These people generally don’t advocate for getting out of the market based on CAPE.

Where CAPE may be useful is for rebalancing and future investments if CAPEs start deviating significantly

Examples:
Reducing international when Japanese CAPE was sky rocketing, increasing US with more reasonable CAPE

US and Intl tracked CAPE well from 1990 until 2009, then deviated substantially (mostly due to valuations), may be a good time to overweight international now as this is one of the largest deltas on record

Value vs TSM is at all time highs with growth having much higher relative valuations
20% VOO | 20% VXUS | 20% AVUV | 20% AVDV | 20% AVES
User avatar
HomerJ
Posts: 21281
Joined: Fri Jun 06, 2008 12:50 pm

Re: CAPE: A much stronger predictor of stock returns than many think

Post by HomerJ »

Nathan Drake wrote: Sun Apr 11, 2021 12:46 pmUS and Intl tracked CAPE well from 1990 until 2009, then deviated substantially (mostly due to valuations), may be a good time to overweight international now as this is one of the largest deltas on record
This is incorrect, at least with regards to the U.S.

CAPE was devised in 1988, and since then has never been accurate at predicting returns for the U.S. stock market.

It's the classic example of a back-test that failed to work going forward.

CAPE went above 20 in 1992, which was "high" historically, which predicted lower than average 10-year returns, and has remained high for the past 29 years (with a brief dip in 2009). Shiller himself predicted 0% real 10-year returns in 1996, and instead we got like 6%-7% real from 1996-2006, basically the historical average.

CAPE has been predicting low 10-year returns going forward since it was invented, and only 2-3 of the 10-year periods were actually low. The rest have been average or even high returns.
"The best tools available to us are shovels, not scalpels. Don't get carried away." - vanBogle59
Robot Monster
Posts: 4215
Joined: Sun May 05, 2019 11:23 am

Re: CAPE: A much stronger predictor of stock returns than many think

Post by Robot Monster »

HomerJ wrote: Sun Apr 11, 2021 1:39 pm CAPE was devised in 1988, and since then has never been accurate at predicting returns for the U.S. stock market.
Here's a post that backs you up written by one of our best forum members,
viewtopic.php?p=4339697#p4339697
Nathan Drake
Posts: 6234
Joined: Mon Apr 11, 2011 12:28 am

Re: CAPE: A much stronger predictor of stock returns than many think

Post by Nathan Drake »

HomerJ wrote: Sun Apr 11, 2021 1:39 pm
Nathan Drake wrote: Sun Apr 11, 2021 12:46 pmUS and Intl tracked CAPE well from 1990 until 2009, then deviated substantially (mostly due to valuations), may be a good time to overweight international now as this is one of the largest deltas on record
This is incorrect, at least with regards to the U.S.

CAPE was devised in 1988, and since then has never been accurate at predicting returns for the U.S. stock market.

It's the classic example of a back-test that failed to work going forward.

CAPE went above 20 in 1992, which was "high" historically, which predicted lower than average 10-year returns, and has remained high for the past 29 years (with a brief dip in 2009). Shiller himself predicted 0% real 10-year returns in 1996, and instead we got like 6%-7% real from 1996-2006, basically the historical average.

CAPE has been predicting low 10-year returns going forward since it was invented, and only 2-3 of the 10-year periods were actually low. The rest have been average or even high returns.
You're misinterpreting what I'm saying. I'm not saying CAPE is a great predictor of what the exact returns will be. But there is a clear relationship between high CAPE values and expected returns. They tend to be higher when CAPE is lower, and they tend to be lower when CAPE is higher. We can't control the investment environment we are currently living in, so "invest we must" is true even if we have lower expected returns. HOWEVER, not all markets have the same CAPE values and sometimes the gap between them can grow quite wide by historical standards. This may imply that other markets are better long-term investments due to lower valuations.

This isn't to say "get out of a certain market entirely" necessarily, but it does mean you may want to expose yourself to different regional diversity and different factor-based investing so you aren't overly exposed to when, say, US TSM experiences a prolonged downturn like 00-09.

"VTSAX and chill" may not be the best advice depending on your end-points of investing to when you start your withdrawal phase. The price you pay for an investment certainly matters.
20% VOO | 20% VXUS | 20% AVUV | 20% AVDV | 20% AVES
User avatar
Topic Author
willthrill81
Posts: 32250
Joined: Thu Jan 26, 2017 2:17 pm
Location: USA
Contact:

Re: CAPE: A much stronger predictor of stock returns than many think

Post by willthrill81 »

Nathan Drake wrote: Sun Apr 11, 2021 3:41 pmBut there is a clear relationship between high CAPE values and expected returns. They tend to be higher when CAPE is lower, and they tend to be lower when CAPE is higher.
Correct, and HomerJ has said so himself in this very thread. But the issue has been that we can't have too much confidence in the CAPE-derived estimate of forward returns. See the below as to why.
marcopolo wrote: Mon Dec 07, 2020 2:39 pm It appears the good correlation comes from the fact that the slope is fairly close, the poor prediction comes from the fact that the intercept is quite different.

The scatter plots, with trend line and equations are shown below:

1881 - 1989:
Image


1989-2010:
Image
I'm quoting yet again marcopolo's excellent post that visually demonstrates how the relationship between CAPE and forward returns was different pre- and post-1989. In both cases, a higher starting CAPE led to lower forward returns. But the intercept (i.e., the expected return when the X variable is equal to zero) was 6% higher after 1989 than before.
Last edited by willthrill81 on Sun Apr 11, 2021 10:05 pm, edited 1 time in total.
The Sensible Steward
User avatar
HomerJ
Posts: 21281
Joined: Fri Jun 06, 2008 12:50 pm

Re: CAPE: A much stronger predictor of stock returns than many think

Post by HomerJ »

Nathan Drake wrote: Sun Apr 11, 2021 3:41 pm
HomerJ wrote: Sun Apr 11, 2021 1:39 pm
Nathan Drake wrote: Sun Apr 11, 2021 12:46 pmUS and Intl tracked CAPE well from 1990 until 2009, then deviated substantially (mostly due to valuations), may be a good time to overweight international now as this is one of the largest deltas on record
This is incorrect, at least with regards to the U.S.

CAPE was devised in 1988, and since then has never been accurate at predicting returns for the U.S. stock market.

It's the classic example of a back-test that failed to work going forward.

CAPE went above 20 in 1992, which was "high" historically, which predicted lower than average 10-year returns, and has remained high for the past 29 years (with a brief dip in 2009). Shiller himself predicted 0% real 10-year returns in 1996, and instead we got like 6%-7% real from 1996-2006, basically the historical average.

CAPE has been predicting low 10-year returns going forward since it was invented, and only 2-3 of the 10-year periods were actually low. The rest have been average or even high returns.
You're misinterpreting what I'm saying. I'm not saying CAPE is a great predictor of what the exact returns will be. But there is a clear relationship between high CAPE values and expected returns. They tend to be higher when CAPE is lower, and they tend to be lower when CAPE is higher.
Sure, but the definition of "higher" has changed over the years. And so have the "expected" returns.

You can always draw a new straight line with the new data, but the old lines were wrong, and the "expected" returns were wrong.

We can't control the investment environment we are currently living in, so "invest we must" is true even if we have lower expected returns. HOWEVER, not all markets have the same CAPE values and sometimes the gap between them can grow quite wide by historical standards. This may imply that other markets are better long-term investments due to lower valuations.
It may imply that, yes. Maybe. Probably not better long-term investments. Maybe better short-term investments.
This isn't to say "get out of a certain market entirely" necessarily, but it does mean you may want to expose yourself to different regional diversity and different factor-based investing so you aren't overly exposed to when, say, US TSM experiences a prolonged downturn like 00-09.

"VTSAX and chill" may not be the best advice depending on your end-points of investing to when you start your withdrawal phase. The price you pay for an investment certainly matters.
Maybe, but trying to avoid the next market downturn may end up making less by increasing exposure to other regions and factor-based investing. All the people on this board who started focusing more on International and factor-based investing in the past 10 years have LESS money than people who ignored valuations.

So far, anyway. Things might certainly change.. But moving money around based on valuations hasn't really worked that well.
"The best tools available to us are shovels, not scalpels. Don't get carried away." - vanBogle59
Nathan Drake
Posts: 6234
Joined: Mon Apr 11, 2011 12:28 am

Re: CAPE: A much stronger predictor of stock returns than many think

Post by Nathan Drake »

HomerJ wrote: Sun Apr 11, 2021 3:55 pm
Nathan Drake wrote: Sun Apr 11, 2021 3:41 pm
HomerJ wrote: Sun Apr 11, 2021 1:39 pm
Nathan Drake wrote: Sun Apr 11, 2021 12:46 pmUS and Intl tracked CAPE well from 1990 until 2009, then deviated substantially (mostly due to valuations), may be a good time to overweight international now as this is one of the largest deltas on record
This is incorrect, at least with regards to the U.S.

CAPE was devised in 1988, and since then has never been accurate at predicting returns for the U.S. stock market.

It's the classic example of a back-test that failed to work going forward.

CAPE went above 20 in 1992, which was "high" historically, which predicted lower than average 10-year returns, and has remained high for the past 29 years (with a brief dip in 2009). Shiller himself predicted 0% real 10-year returns in 1996, and instead we got like 6%-7% real from 1996-2006, basically the historical average.

CAPE has been predicting low 10-year returns going forward since it was invented, and only 2-3 of the 10-year periods were actually low. The rest have been average or even high returns.
You're misinterpreting what I'm saying. I'm not saying CAPE is a great predictor of what the exact returns will be. But there is a clear relationship between high CAPE values and expected returns. They tend to be higher when CAPE is lower, and they tend to be lower when CAPE is higher.
Sure, but the definition of "higher" has changed over the years. And so have the "expected" returns.

You can always draw a new straight line with the new data, but the old lines were wrong, and the "expected" returns were wrong.

We can't control the investment environment we are currently living in, so "invest we must" is true even if we have lower expected returns. HOWEVER, not all markets have the same CAPE values and sometimes the gap between them can grow quite wide by historical standards. This may imply that other markets are better long-term investments due to lower valuations.
It may imply that, yes. Maybe. Probably not better long-term investments. Maybe better short-term investments.
This isn't to say "get out of a certain market entirely" necessarily, but it does mean you may want to expose yourself to different regional diversity and different factor-based investing so you aren't overly exposed to when, say, US TSM experiences a prolonged downturn like 00-09.

"VTSAX and chill" may not be the best advice depending on your end-points of investing to when you start your withdrawal phase. The price you pay for an investment certainly matters.
Maybe, but trying to avoid the next market downturn may end up making less by increasing exposure to other regions and factor-based investing. All the people on this board who started focusing more on International and factor-based investing in the past 10 years have LESS money than people who ignored valuations.

So far, anyway. Things might certainly change.. But moving money around based on valuations hasn't really worked that well.
The gap between growth / value and US / International hasn't manifested itself until recently. Again, the period of 00-09 was one where US TSM was one of the poorest investment classes to be in, and where International and Value did quite well. A "dead decade" is something many of us would like to avoid. Yes, during some investment periods such as the past decade, you are likely to underperform whatever is the best asset class in your portfolio. But that is merely a consequence of having a more diverse portfolio.

The fact that we don't know, or can't time, when these "regime shifts" occur, is more evidence that it's probably a good idea to invest in them at all times and never be 100% in or out of any particular geographical area or risk factor.

Moving money based upon valuations is what you would expect through normal re-balancing. Same thing occurs with stocks/vs bonds will occur with US vs International vs other Factors.
20% VOO | 20% VXUS | 20% AVUV | 20% AVDV | 20% AVES
Barsoom
Posts: 728
Joined: Thu Dec 06, 2018 8:40 am

Re: CAPE: A much stronger predictor of stock returns than many think

Post by Barsoom »

Hydromod wrote: Sun Jan 10, 2021 4:57 pm Just for grins, I updated the related approach from Philosophical Economist with data since I last looked at it in 2019. This theory was mentioned upthread here.
Thank you for following up on my post.

-B
Barsoom
Posts: 728
Joined: Thu Dec 06, 2018 8:40 am

Re: CAPE: A much stronger predictor of stock returns than many think

Post by Barsoom »

klaus14 wrote: Sun Jan 10, 2021 6:57 pmLatest data from FRED seems to be Q3 alloc = 45%.
Since then market went up. I'd guess it's like 47% now, which is the highest in history except Q1 2000 when it was 52%
The data from your link IS actually 47.6%.

-B
User avatar
HomerJ
Posts: 21281
Joined: Fri Jun 06, 2008 12:50 pm

Re: CAPE: A much stronger predictor of stock returns than many think

Post by HomerJ »

Nathan Drake wrote: Sun Apr 11, 2021 4:01 pm The fact that we don't know, or can't time, when these "regime shifts" occur, is more evidence that it's probably a good idea to invest in them at all times and never be 100% in or out of any particular geographical area or risk factor.

Moving money based upon valuations is what you would expect through normal re-balancing. Same thing occurs with stocks/vs bonds will occur with US vs International vs other Factors.
True and true.
"The best tools available to us are shovels, not scalpels. Don't get carried away." - vanBogle59
klaus14
Posts: 953
Joined: Sun Nov 25, 2018 6:43 pm

Re: CAPE: A much stronger predictor of stock returns than many think

Post by klaus14 »

Barsoom wrote: Sun Apr 11, 2021 4:05 pm
klaus14 wrote: Sun Jan 10, 2021 6:57 pmLatest data from FRED seems to be Q3 alloc = 45%.
Since then market went up. I'd guess it's like 47% now, which is the highest in history except Q1 2000 when it was 52%
The data from your link IS actually 47.6%.

-B
Yes. They updated for Q4 after my post.
My investment algorithm: https://www.bogleheads.org/forum/viewtopic.php?f=10&t=351899&p=6112869#p6112869
tvubpwcisla
Posts: 1167
Joined: Sat Nov 09, 2019 9:09 am

Re: CAPE: A much stronger predictor of stock returns than many think

Post by tvubpwcisla »

I thought CAPE stood for:

Careful
Analysts
Pick
Equities

oh, how things have changed!
inbox788
Posts: 8372
Joined: Thu Mar 15, 2012 5:24 pm

Re: CAPE: A much stronger predictor of stock returns than many think

Post by inbox788 »

Just looked at the current PE ratios. What is CAPE predicting about future returns? How far into the future and what direction?

The red line on the 1989-2010 scatterplot looks about negative 2% when CAPE is 37, correct?

Current S&P 500 PE Ratio: 42.00 -0.04 (-0.09%)
1:29 PM EDT, Mon Apr 12
https://www.multpl.com/s-p-500-pe-ratio

Current Shiller PE Ratio: 37.06 -0.03 (-0.09%)
1:29 PM EDT, Mon Apr 12
https://www.multpl.com/shiller-pe
This ratio was at a record 28 in January 1997, with the only other instance (at that time) of a comparably high ratio occurring in 1929. Shiller and Campbell asserted the ratio was predicting that the real value of the market would be 40% lower in ten years than it was at that time. That forecast proved to be remarkably prescient, as the market crash of 2008 contributed to the S&P 500 plunging 60% from October 2007 to March 2009.
https://www.investopedia.com/terms/c/cape-ratio.asp
Dirghatamas
Posts: 583
Joined: Fri Jan 01, 2016 5:18 pm

Re: CAPE: A much stronger predictor of stock returns than many think

Post by Dirghatamas »

YRT70 wrote: Sat Apr 10, 2021 12:54 pm Perhaps interesting to some here, Ben Felix recently commented on the predictive power of CAPE on his forum.
Basically, the predictive power of CAPE goes away if you consider independent samples (non overlapping) instead of rolling periods. It wasn’t this paper that changed my mind (it was a long talk with someone from Dimensional’s research group) but this paper touches the same points.

https://papers.ssrn.com/sol3/papers.cfm ... id=3142575
Thankyou YRT70 for the link to a brilliant paper! I had always worried about how people (including Shiller) were using overlapping samples to create enough data points for statistical techniques to make sense but then using an averaging technique (what cape is) as its input...that made no sense to me because the points are not independent by a long shot.

This paper does a brilliant job of formalizing the vague issue many of us may have had into hard analysis. Regardless of what one may feel about CAPE (is it high, low, predictive or not), I would recommend reading the paper because according to the paper, the CAPE statistical analysis technique often used to verify if it is a good predictor, itself doesn't make sense.
marcopolo
Posts: 8445
Joined: Sat Dec 03, 2016 9:22 am

Re: CAPE: A much stronger predictor of stock returns than many think

Post by marcopolo »

inbox788 wrote: Mon Apr 12, 2021 12:38 pm Just looked at the current PE ratios. What is CAPE predicting about future returns? How far into the future and what direction?

The red line on the 1989-2010 scatterplot looks about negative 2% when CAPE is 37, correct?

Current S&P 500 PE Ratio: 42.00 -0.04 (-0.09%)
1:29 PM EDT, Mon Apr 12
https://www.multpl.com/s-p-500-pe-ratio

Current Shiller PE Ratio: 37.06 -0.03 (-0.09%)
1:29 PM EDT, Mon Apr 12
https://www.multpl.com/shiller-pe
This ratio was at a record 28 in January 1997, with the only other instance (at that time) of a comparably high ratio occurring in 1929. Shiller and Campbell asserted the ratio was predicting that the real value of the market would be 40% lower in ten years than it was at that time. That forecast proved to be remarkably prescient, as the market crash of 2008 contributed to the S&P 500 plunging 60% from October 2007 to March 2009.
https://www.investopedia.com/terms/c/cape-ratio.asp
The point of those plots was to show that the relationship between CAPE and forward returns is non-stationary. That is the relationship changes over time, and what may look like a good correlation can lead to poor predictions.

For example, looking at data prior to 1989, that same CAPE = 37 would have predicted something around -8%, and that is only by extrapolating the line because CAPE had NEVER been that high before.

People could have looked at the pre-1989 data and suggested that a CAPE of ~27 would lead to -2%. But, in the post 1989 plot it looks more like ~3%, and it took a CAPE value a full 10 points higher to achieve the same predicted return!

There are too many things that change like accounting rules (GAAP reporting changes), composition of economy (higher growth tech industry vs. slower growth utilities), etc. to think that this relationship should stay consistent. Historical data shows it has not. There is no reason to think it will in the future.
Once in a while you get shown the light, in the strangest of places if you look at it right.
User avatar
HomerJ
Posts: 21281
Joined: Fri Jun 06, 2008 12:50 pm

Re: CAPE: A much stronger predictor of stock returns than many think

Post by HomerJ »

inbox788 wrote: Mon Apr 12, 2021 12:38 pm
This ratio was at a record 28 in January 1997, with the only other instance (at that time) of a comparably high ratio occurring in 1929. Shiller and Campbell asserted the ratio was predicting that the real value of the market would be 40% lower in ten years than it was at that time. That forecast proved to be remarkably prescient, as the market crash of 2008 contributed to the S&P 500 plunging 60% from October 2007 to March 2009.
https://www.investopedia.com/terms/c/cape-ratio.asp
Look guys... You can't automatically trust anything you read about the stock market. Anyone can twist the data to make their points.

What's really bad is when they don't even TRY to twist the data. They just assert something and you just believe it.

Some of this stuff is easy to check.

Notice that the quote above talks about how the S&P "plunged 60%" from Oct 2007 - March 2009, but it doesn't talk about how much it gained before that.

So let's test how "remarkably prescient" Shiller's CAPE predictions were.

Shiller asserted in Jan 1997 that the real value of the market would be 40% lower in ten years.

Here's the chart from 1997-2007

Image

At NO point was the investor ever even DOWN during the ten years in real terms (although close at the bottom in 2003), and at the end of the ten years, the investor had almost doubled their money in real terms. The $10,000 grew to around $18,000 (inflation adjusted), it had grown to $22,763 nominal.

Investing in 1997, highest valuations in history at the time, was actually a decent time to invest.

Does that sound like Shiller was "remarkably prescient?"

Okay, let's add a few extra years to give Shiller a chance...

Here's the chart from 1997 through 2010

Image

Yes, there was a huge crash from 2007 to 2009. Was the investor ever 40% down in real terms?

Nope, the increase in the years before 2007 almost completely shielded the investor from the 2009 crash. There was a couple of weeks where the investor was down about 7% in real terms at $9,300 (still up at $12,400 in nominal terms). By Dec 31st, 2009 the investor was back to $14,290 in real terms ($19,457 in nominal).

So was Shiller "remarkably prescient"?

Was his amazing predictive powers actionable? Could one have gotten out of the market in Jan 1997, and made a bunch more money waiting for the 40% real crash from 1997 levels? No... It never happened.

Read the stuff carefully. Learn the actual history. Look up the actual predictions from the past. They are still there on the Internet. When someone says CAPE is super-predictive, it's not hard to find actual predictions from the past written by Nobel-prize winning economists using CAPE models that failed to accurately predict stock market movements.

Use portfolio visualizer to see if the numbers are real.
"The best tools available to us are shovels, not scalpels. Don't get carried away." - vanBogle59
Barsoom
Posts: 728
Joined: Thu Dec 06, 2018 8:40 am

Re: CAPE: A much stronger predictor of stock returns than many think

Post by Barsoom »

Random Walker wrote: Sat Apr 10, 2021 10:21 pm
HootingSloth wrote: Sat Apr 10, 2021 9:03 pm
Random Walker wrote: Sat Apr 10, 2021 10:11 am A Monte Carlo Simulation could be very useful for you in making the decisions. I’m guessing you might be surprised by the lack of sensitivity the likelihood of meeting your goals is to asset allocation. You may actually find less aggressive allocations are more likely to achieve your goals while resulting in lower terminal value. More aggressive allocation will result in higher average terminal value, but significantly wider dispersion of potential outcomes.

Dave
Dave,

I realized that I had never really done this myself despite spending a lot of time playing around with Portfolio Visualizer. This was an interesting exercise for me as well, and I'm glad you suggested it.

I was somewhat surprised how little of a difference asset allocation made in our circumstances, even switching between 80/20 and 20/80 over the span of multiple decades. Obviously, the "good" tail was pretty different, but a lot of the results were more in the "Would I really notice this difference?" range than I expected. 80/20 went from "That seems like enough" to "Maybe I should start reading about estate planning and NetJets" while 20/80 went from "I won't worry about whether this is enough" to "This will be a very luxurious lifestyle." And that's even with the aggressively fat tails that you get from Monte Carlo simulations.

It was helpful in reaffirming that a lot of these decisions are less important than just avoiding some big behavioral mistake.
I’ve found MCS to be enlightening. Like I said, I was surprised to see the lack of sensitivity of results to the overall asset allocation. Implicit in the exercise, it really focuses the investor on clarifying needs versus wants. Ive found that focus valuable. I guess in the financial advisor business they divide goals into needs, wants, and wishes.

Avoiding big behavioral mistakes is huge. I ultimately evolved from a DIY TSM Boglehead to a Factorhead using an advisor. I did a lot of spreadsheets in making the decision, and I even tried to factor in avoidance of behavioral errors. May have helped me arrive at the result I was looking for, but I thought avoidance of behavioral errors dominated expense ratio differences. A couple or few behavioral errors
in an investing career can make the expense ratio differences we argue about here look silly.

A lot of people here talk about cooling off their portfolio from 80/20 to 70/30 or 60/40 in retirement. When even a 60/40 TSM portfolio has more than 85% of its risk wrapped up in the market factor, I don’t think they are cooling off the portfolio as much as they think. It’s worthwhile to use MCS to really look at how much or how little bigger differences in the basic AA, as you did from 20/80 to 80/20, affect outcomes. The trade off for pretty well off individuals between mean terminal value and likelihood of not outliving money is eye opening.

Dave
Let me offer a different perspective.

I am retired for just under two years and so am in the decumulation phase. In my retirement portfolio, 90% of it is before-tax, so all of my withdrawals are taxed at ordinary income rates. I have no ROTH IRAs. Furthermore, at some point in the future social security will become a source of additional income. Finally, RMDs will force me to withdraw more than I need to meet expenses, which will begin to grow my after-tax balance. Earnings in the after-tax account will be subject to taxes.

For me, my goal is to find the highest safe withdrawal rate that will last for 30 years. I don't have children, so I am not looking to leave behind a legacy.

I use the BH Retiree Portfolio Model to forecast my cash flow. I also wrote a Monte Carlo variant to the RPM that allows me to add uncertainty and sequence of return risk. Furthermore, I can run the Monte Carlo simulations from within Excel's Solver so that it can adjust my asset allocations and withdrawal rates to maximize my living expenses and still remain within a 5% probability of running out of money (my risk tolerance). For the Solver, I use OpenSolver.org's NOMAD non-linear solver.

I am using the "Age Banding" approach to adjust my living expenses over time.

I ran two scenarios for Solver to optimize.

Scenario #1: Minimize final portfolio value

My assumptions in this scenario are:
  1. Increasing my living expense withdrawals (and associated taxes) will draw down my balance, and this will lead to a lower final portfolio balance.
  2. Adjusting my AA to "safe" investments will reduce the risk of losses, which will leave more to spend.
Scenario #2: Maximize annual withdrawals

My assumptions in this scenario are:
  1. Increasing my living expense withdrawals (and associated taxes) will draw down my balance, and this will lead to a lower final portfolio balance.
  2. Adjusting my AA to "riskier" investments will increase the chance of future gains, which will leave more to spend.
Here is what I found using my portfolio.

Results for Scenario #1: Minimize final portfolio value
  1. Before-Tax AA: stocks: 2.2%, bonds: 65.7%, money market/cash: 32.1%.
  2. After-Tax AA: stocks: 15.3%, bonds: 4.2%, money market/cash: 80.5%.
  3. Aggregate AA: stocks: 3.4%, bonds: 59.7%, money market/cash: 36.9%.
  4. The Expected Value (probability weighted average) of the final portfolio value was 34.5% of my starting balance at retirement. One-third of my portfolio was left behind.
  5. The P10 (10% probability point) was 10.5% of my starting balance.
  6. The P50 was 35.8% of my starting balance.
  7. The P90 was 56.5% of my starting balance.
  8. My maximum living expenses (before taxes) in year 1 was 3.8% of my starting balance.
    • This is 25.5% above my actual before-taxes living expenses.

Scenario #2: Maximize annual withdrawals
  1. Before-Tax AA: stocks: 67.0%, bonds: 31.9%, money market/cash: 1.1%.
  2. After-Tax AA: stocks: 37,6%, bonds: 58.3%, money market/cash: 4.1%.
  3. Aggregate AA: stocks: 64.2%, bonds: 34.4%, money market/cash: 1.4%.
  4. The Expected Value of the final portfolio value was 163.0% of my starting balance at retirement. My final portfolio balance was was more than 50% greater than when I started.
  5. The P10 was 22.8% of my starting balance.
  6. The P50 was 127.1% of my starting balance.
  7. The P90 was 347.1% of my starting balance.
  8. My maximum living expenses in year 1 was 4.3% of my starting balance.
    • This is 41.8% above my actual before-taxes living expenses.
Conclusions
  1. Going with a 3%-60%-37% AA (bond/cash) resulted in a smaller swing in the low/high range of outcomes.
    • It also resulted in a lifestyle that was 11.5% smaller than the stock/bond allocation affords.
  2. Going with the 64%-34%-1% AA (stock/bond) resulted in a wide swing in outcomes.
    • It also resulted in a lifestyle that was 13.0% higher than the bond/cash allocation pays for.
    • The portfolio also was left with more money in it than it began with.
    • The wide swing in outcomes due to majority of equities is resulting in leaving money "off the table" for expenses in order to protect against running out of money.
  3. I would assume that I would redo the analysis each year as another year of "risk" becomes sunk. This would result in an increasing withdrawal rate for the remaining years as the risk horizon reduces. Hopefully, this will end up with less terminal value than is assumed at year 1 of retirement.
With the caveat that this is based on my own home-grown Monte Carlo simulator, if you are seeing reduced sensitivities due to asset allocation during the accumulation phase, that is not the case during the decumulation phase.

-B
User avatar
Nicolas Perrault
Posts: 244
Joined: Thu Sep 27, 2018 12:33 pm
Location: Oxford, UK

Re: U.S. stocks continue to soar!

Post by Nicolas Perrault »

[Moved into a new thread from: U.S. stocks continue to soar! --admin LadyGeek]
squirm wrote: Fri Nov 05, 2021 1:34 pm Going back to the roaring twenties, the Dow advanced about 4-5x from the earlier low.

Given a similar situation, I think we could easily see Dow 100,000-120,000, S&P 15,000-20,000 and Nasdaq 50,000 by 2027/2028.
The P/E 10 (Shiller PE) in 1920 was 4.78 --- the all-time record low (1871-2021 data). This suggested forward returns of 1/4.78 = 20.9% (= earnings yield). Over the next 8.75 years, including dividends, your investment between Dec 1920 and Sep 1929 returned a total real 27.2% annualized (x8.2, using the quotient of Shiller's Total Return Prices for the two dates).

Today the P/E 10 is of 40.00 (source), the second highest it has ever been (after 1999-2000). This suggests forward returns of 1/40 = 2.5%. That's the strength of the price "motor" inside the S&P500 nowadays. Quite different from the price motor in 1920.

For the markets to do x5 in 6-7 years, the P/E 10 would need to reach almost 200. "Given a similar situation..." Let me stop you right there, it's not a similar situation.
User avatar
HomerJ
Posts: 21281
Joined: Fri Jun 06, 2008 12:50 pm

Re: U.S. stocks continue to soar!

Post by HomerJ »

Nicolas Perrault wrote: Fri Nov 05, 2021 7:14 pm
squirm wrote: Fri Nov 05, 2021 1:34 pm Going back to the roaring twenties, the Dow advanced about 4-5x from the earlier low.

Given a similar situation, I think we could easily see Dow 100,000-120,000, S&P 15,000-20,000 and Nasdaq 50,000 by 2027/2028.
The P/E 10 (Shiller PE) in 1920 was 4.78 --- the all-time record low (1871-2021 data). This suggested forward returns of 1/4.78 = 20.9% (= earnings yield). Over the next 8.75 years, including dividends, your investment between Dec 1920 and Sep 1929 returned a total real 27.2% annualized (x8.2, using the quotient of Shiller's Total Return Prices for the two dates).

Today the P/E 10 is of 40.00 (source), the second highest it has ever been (after 1999-2000). This suggests forward returns of 1/40 = 2.5%. That's the strength of the price "motor" inside the S&P500 nowadays. Quite different from the price motor in 1920.

For the markets to do x5 in 6-7 years, the P/E 10 would need to reach almost 200. "Given a similar situation..." Let me stop you right there, it's not a similar situation.
1/CAPE was never a thing until about 5-10 years ago, when people noticed that 1/CAPE actually modeled the new line.

For instance, in 1996, a few years after Shiller INVENTED CAPE, when CAPE was 25, he predicted 0% real return, not 4% real.

Because that is what the data said.

The data has changed over the past 30 years, and NOW, 1/CAPE is somewhat close to the NEW line, but don't fool yourself that there is some cosmic law that says 1/CAPE actually has predictive power.
"The best tools available to us are shovels, not scalpels. Don't get carried away." - vanBogle59
User avatar
Nicolas Perrault
Posts: 244
Joined: Thu Sep 27, 2018 12:33 pm
Location: Oxford, UK

Re: U.S. stocks continue to soar!

Post by Nicolas Perrault »

HomerJ wrote: Fri Nov 05, 2021 7:28 pm
Nicolas Perrault wrote: Fri Nov 05, 2021 7:14 pm
squirm wrote: Fri Nov 05, 2021 1:34 pm Going back to the roaring twenties, the Dow advanced about 4-5x from the earlier low.

Given a similar situation, I think we could easily see Dow 100,000-120,000, S&P 15,000-20,000 and Nasdaq 50,000 by 2027/2028.
The P/E 10 (Shiller PE) in 1920 was 4.78 --- the all-time record low (1871-2021 data). This suggested forward returns of 1/4.78 = 20.9% (= earnings yield). Over the next 8.75 years, including dividends, your investment between Dec 1920 and Sep 1929 returned a total real 27.2% annualized (x8.2, using the quotient of Shiller's Total Return Prices for the two dates).

Today the P/E 10 is of 40.00 (source), the second highest it has ever been (after 1999-2000). This suggests forward returns of 1/40 = 2.5%. That's the strength of the price "motor" inside the S&P500 nowadays. Quite different from the price motor in 1920.

For the markets to do x5 in 6-7 years, the P/E 10 would need to reach almost 200. "Given a similar situation..." Let me stop you right there, it's not a similar situation.
1/CAPE was never a thing until about 5-10 years ago, when people noticed that 1/CAPE actually modeled the new line.

For instance, in 1996, a few years after Shiller INVENTED CAPE, when CAPE was 25, he predicted 0% real return, not 4% real.

Because that is what the data said.

The data has changed over the past 30 years, and NOW, 1/CAPE is somewhat close to the NEW line, but don't fool yourself that there is some cosmic law that says 1/CAPE actually has predictive power.
I'm not sure I follow you HomerJ. What "new line" are you talking about? The subsequent annualized 10-year excess return line?

When Shiller predicted 0% real return in 1996, presumably his model assumed some form of regression to the mean for the CAPE. But the return projections of 1/CAPE do not.

Ain't nobody who talked about no cosmic law for 1/CAPE. But who can deny that stocks are more attractive with a CAPE of 5 than one of 40? The CAPE is simply the price you pay for $1 of earnings. Would you rather pay $5 or $40 for that $1 of earnings? Would you rather pay $5 for a hamburger or $40?
User avatar
Topic Author
willthrill81
Posts: 32250
Joined: Thu Jan 26, 2017 2:17 pm
Location: USA
Contact:

Re: U.S. stocks continue to soar!

Post by willthrill81 »

Nicolas Perrault wrote: Fri Nov 05, 2021 7:46 pm
HomerJ wrote: Fri Nov 05, 2021 7:28 pm
Nicolas Perrault wrote: Fri Nov 05, 2021 7:14 pm
squirm wrote: Fri Nov 05, 2021 1:34 pm Going back to the roaring twenties, the Dow advanced about 4-5x from the earlier low.

Given a similar situation, I think we could easily see Dow 100,000-120,000, S&P 15,000-20,000 and Nasdaq 50,000 by 2027/2028.
The P/E 10 (Shiller PE) in 1920 was 4.78 --- the all-time record low (1871-2021 data). This suggested forward returns of 1/4.78 = 20.9% (= earnings yield). Over the next 8.75 years, including dividends, your investment between Dec 1920 and Sep 1929 returned a total real 27.2% annualized (x8.2, using the quotient of Shiller's Total Return Prices for the two dates).

Today the P/E 10 is of 40.00 (source), the second highest it has ever been (after 1999-2000). This suggests forward returns of 1/40 = 2.5%. That's the strength of the price "motor" inside the S&P500 nowadays. Quite different from the price motor in 1920.

For the markets to do x5 in 6-7 years, the P/E 10 would need to reach almost 200. "Given a similar situation..." Let me stop you right there, it's not a similar situation.
1/CAPE was never a thing until about 5-10 years ago, when people noticed that 1/CAPE actually modeled the new line.

For instance, in 1996, a few years after Shiller INVENTED CAPE, when CAPE was 25, he predicted 0% real return, not 4% real.

Because that is what the data said.

The data has changed over the past 30 years, and NOW, 1/CAPE is somewhat close to the NEW line, but don't fool yourself that there is some cosmic law that says 1/CAPE actually has predictive power.
I'm not sure I follow you HomerJ. What "new line" are you talking about? The subsequent annualized 10-year excess return line?

When Shiller predicted 0% real return in 1996, presumably his model assumed some form of regression to the mean for the CAPE. But the return projections of 1/CAPE do not.

Ain't nobody who talked about no cosmic law for 1/CAPE. But who can deny that stocks are more attractive with a CAPE of 5 than one of 40? The CAPE is simply the price you pay for $1 of earnings. Would you rather pay $5 or $40 for that $1 of earnings? Would you rather pay $5 for a hamburger or $40?
I started a thread that became very long in which we discussed this issue. HomerJ posted his views in it often. It might be better to move this discussion into that thread to avoid derailing this one.
The Sensible Steward
nigel_ht
Posts: 4742
Joined: Tue Jan 01, 2019 9:14 am

Re: U.S. stocks continue to soar!

Post by nigel_ht »

Nicolas Perrault wrote: Fri Nov 05, 2021 7:46 pm
HomerJ wrote: Fri Nov 05, 2021 7:28 pm
Nicolas Perrault wrote: Fri Nov 05, 2021 7:14 pm
squirm wrote: Fri Nov 05, 2021 1:34 pm Going back to the roaring twenties, the Dow advanced about 4-5x from the earlier low.

Given a similar situation, I think we could easily see Dow 100,000-120,000, S&P 15,000-20,000 and Nasdaq 50,000 by 2027/2028.
The P/E 10 (Shiller PE) in 1920 was 4.78 --- the all-time record low (1871-2021 data). This suggested forward returns of 1/4.78 = 20.9% (= earnings yield). Over the next 8.75 years, including dividends, your investment between Dec 1920 and Sep 1929 returned a total real 27.2% annualized (x8.2, using the quotient of Shiller's Total Return Prices for the two dates).

Today the P/E 10 is of 40.00 (source), the second highest it has ever been (after 1999-2000). This suggests forward returns of 1/40 = 2.5%. That's the strength of the price "motor" inside the S&P500 nowadays. Quite different from the price motor in 1920.

For the markets to do x5 in 6-7 years, the P/E 10 would need to reach almost 200. "Given a similar situation..." Let me stop you right there, it's not a similar situation.
1/CAPE was never a thing until about 5-10 years ago, when people noticed that 1/CAPE actually modeled the new line.

For instance, in 1996, a few years after Shiller INVENTED CAPE, when CAPE was 25, he predicted 0% real return, not 4% real.

Because that is what the data said.

The data has changed over the past 30 years, and NOW, 1/CAPE is somewhat close to the NEW line, but don't fool yourself that there is some cosmic law that says 1/CAPE actually has predictive power.
I'm not sure I follow you HomerJ. What "new line" are you talking about? The subsequent annualized 10-year excess return line?

When Shiller predicted 0% real return in 1996, presumably his model assumed some form of regression to the mean for the CAPE. But the return projections of 1/CAPE do not.

Ain't nobody who talked about no cosmic law for 1/CAPE. But who can deny that stocks are more attractive with a CAPE of 5 than one of 40? The CAPE is simply the price you pay for $1 of earnings. Would you rather pay $5 or $40 for that $1 of earnings? Would you rather pay $5 for a hamburger or $40?
He’s saying as the data changed so did the model to match.

So while the current model is a good fit for today’s data if you look at the track record of the prior models…maybe not so much a great predictor of future performance.
nigel_ht
Posts: 4742
Joined: Tue Jan 01, 2019 9:14 am

Re: U.S. stocks continue to soar!

Post by nigel_ht »

willthrill81 wrote: Fri Nov 05, 2021 8:00 pm
Nicolas Perrault wrote: Fri Nov 05, 2021 7:46 pm
HomerJ wrote: Fri Nov 05, 2021 7:28 pm
Nicolas Perrault wrote: Fri Nov 05, 2021 7:14 pm
squirm wrote: Fri Nov 05, 2021 1:34 pm Going back to the roaring twenties, the Dow advanced about 4-5x from the earlier low.

Given a similar situation, I think we could easily see Dow 100,000-120,000, S&P 15,000-20,000 and Nasdaq 50,000 by 2027/2028.
The P/E 10 (Shiller PE) in 1920 was 4.78 --- the all-time record low (1871-2021 data). This suggested forward returns of 1/4.78 = 20.9% (= earnings yield). Over the next 8.75 years, including dividends, your investment between Dec 1920 and Sep 1929 returned a total real 27.2% annualized (x8.2, using the quotient of Shiller's Total Return Prices for the two dates).

Today the P/E 10 is of 40.00 (source), the second highest it has ever been (after 1999-2000). This suggests forward returns of 1/40 = 2.5%. That's the strength of the price "motor" inside the S&P500 nowadays. Quite different from the price motor in 1920.

For the markets to do x5 in 6-7 years, the P/E 10 would need to reach almost 200. "Given a similar situation..." Let me stop you right there, it's not a similar situation.
1/CAPE was never a thing until about 5-10 years ago, when people noticed that 1/CAPE actually modeled the new line.

For instance, in 1996, a few years after Shiller INVENTED CAPE, when CAPE was 25, he predicted 0% real return, not 4% real.

Because that is what the data said.

The data has changed over the past 30 years, and NOW, 1/CAPE is somewhat close to the NEW line, but don't fool yourself that there is some cosmic law that says 1/CAPE actually has predictive power.
I'm not sure I follow you HomerJ. What "new line" are you talking about? The subsequent annualized 10-year excess return line?

When Shiller predicted 0% real return in 1996, presumably his model assumed some form of regression to the mean for the CAPE. But the return projections of 1/CAPE do not.

Ain't nobody who talked about no cosmic law for 1/CAPE. But who can deny that stocks are more attractive with a CAPE of 5 than one of 40? The CAPE is simply the price you pay for $1 of earnings. Would you rather pay $5 or $40 for that $1 of earnings? Would you rather pay $5 for a hamburger or $40?
I started a thread that became very long in which we discussed this issue. HomerJ posted his views in it often. It might be better to move this discussion into that thread to avoid derailing this one.
I hate to say that my primary take away from that thread was that Forester is actually a reasonable poster except when yanking people’s chains in this thread…which is pretty much what this thread is for anyway…BHers shouldn’t care about either soaring or freefalling….
User avatar
HomerJ
Posts: 21281
Joined: Fri Jun 06, 2008 12:50 pm

Re: U.S. stocks continue to soar!

Post by HomerJ »

Nicolas Perrault wrote: Fri Nov 05, 2021 7:46 pm
HomerJ wrote: Fri Nov 05, 2021 7:28 pm
Nicolas Perrault wrote: Fri Nov 05, 2021 7:14 pm
squirm wrote: Fri Nov 05, 2021 1:34 pm Going back to the roaring twenties, the Dow advanced about 4-5x from the earlier low.

Given a similar situation, I think we could easily see Dow 100,000-120,000, S&P 15,000-20,000 and Nasdaq 50,000 by 2027/2028.
The P/E 10 (Shiller PE) in 1920 was 4.78 --- the all-time record low (1871-2021 data). This suggested forward returns of 1/4.78 = 20.9% (= earnings yield). Over the next 8.75 years, including dividends, your investment between Dec 1920 and Sep 1929 returned a total real 27.2% annualized (x8.2, using the quotient of Shiller's Total Return Prices for the two dates).

Today the P/E 10 is of 40.00 (source), the second highest it has ever been (after 1999-2000). This suggests forward returns of 1/40 = 2.5%. That's the strength of the price "motor" inside the S&P500 nowadays. Quite different from the price motor in 1920.

For the markets to do x5 in 6-7 years, the P/E 10 would need to reach almost 200. "Given a similar situation..." Let me stop you right there, it's not a similar situation.
1/CAPE was never a thing until about 5-10 years ago, when people noticed that 1/CAPE actually modeled the new line.

For instance, in 1996, a few years after Shiller INVENTED CAPE, when CAPE was 25, he predicted 0% real return, not 4% real.

Because that is what the data said.

The data has changed over the past 30 years, and NOW, 1/CAPE is somewhat close to the NEW line, but don't fool yourself that there is some cosmic law that says 1/CAPE actually has predictive power.
I'm not sure I follow you HomerJ. What "new line" are you talking about? The subsequent annualized 10-year excess return line?
The new line of best fit has changed over the years. 1/CAPE now approximates that line. It didn't in the past. You wrote an entire post using 1/CAPE to calculate forward returns.

1/CAPE didn't fit the data in the past. That's a recent (and likely random) new development.
"The best tools available to us are shovels, not scalpels. Don't get carried away." - vanBogle59
User avatar
Nicolas Perrault
Posts: 244
Joined: Thu Sep 27, 2018 12:33 pm
Location: Oxford, UK

Re: U.S. stocks continue to soar!

Post by Nicolas Perrault »

HomerJ wrote: Fri Nov 05, 2021 11:07 pm
Nicolas Perrault wrote: Fri Nov 05, 2021 7:46 pm
HomerJ wrote: Fri Nov 05, 2021 7:28 pm
Nicolas Perrault wrote: Fri Nov 05, 2021 7:14 pm
squirm wrote: Fri Nov 05, 2021 1:34 pm Going back to the roaring twenties, the Dow advanced about 4-5x from the earlier low.

Given a similar situation, I think we could easily see Dow 100,000-120,000, S&P 15,000-20,000 and Nasdaq 50,000 by 2027/2028.
The P/E 10 (Shiller PE) in 1920 was 4.78 --- the all-time record low (1871-2021 data). This suggested forward returns of 1/4.78 = 20.9% (= earnings yield). Over the next 8.75 years, including dividends, your investment between Dec 1920 and Sep 1929 returned a total real 27.2% annualized (x8.2, using the quotient of Shiller's Total Return Prices for the two dates).

Today the P/E 10 is of 40.00 (source), the second highest it has ever been (after 1999-2000). This suggests forward returns of 1/40 = 2.5%. That's the strength of the price "motor" inside the S&P500 nowadays. Quite different from the price motor in 1920.

For the markets to do x5 in 6-7 years, the P/E 10 would need to reach almost 200. "Given a similar situation..." Let me stop you right there, it's not a similar situation.
1/CAPE was never a thing until about 5-10 years ago, when people noticed that 1/CAPE actually modeled the new line.

For instance, in 1996, a few years after Shiller INVENTED CAPE, when CAPE was 25, he predicted 0% real return, not 4% real.

Because that is what the data said.

The data has changed over the past 30 years, and NOW, 1/CAPE is somewhat close to the NEW line, but don't fool yourself that there is some cosmic law that says 1/CAPE actually has predictive power.
I'm not sure I follow you HomerJ. What "new line" are you talking about? The subsequent annualized 10-year excess return line?
The new line of best fit has changed over the years. 1/CAPE now approximates that line. It didn't in the past. You wrote an entire post using 1/CAPE to calculate forward returns.

1/CAPE didn't fit the data in the past. That's a recent (and likely random) new development.
I have no idea what line of best fit you‘re talking about. Nor have I claimed to have „calculated“ forward returns. No one can do that.

I see from your post history that you have a strong aversion to the CAPE metric because you feel that people have been using it to „predict the past“.You also seem to dislike the argument that „the CAPE is high so it must fall“. Check my post, I did not say this.

What I do say is this. If the CAPE is at 5 today and stays at 5 for the foreseeable future, you are far more likely to get greater returns over the next decade than if it is now at 40 and stays at 40 for the foreseeable future. To produce greater returns in the latter scenario, the earnings growth would have to be massive to make up for the tiny initial earnings yield. You cannot escape this HomerJ, it‘s just math.
Soysauceonrice
Posts: 55
Joined: Thu Aug 05, 2021 9:39 pm

Re: U.S. stocks continue to soar!

Post by Soysauceonrice »

Nicolas Perrault wrote: Fri Nov 05, 2021 11:49 pm
HomerJ wrote: Fri Nov 05, 2021 11:07 pm
Nicolas Perrault wrote: Fri Nov 05, 2021 7:46 pm
HomerJ wrote: Fri Nov 05, 2021 7:28 pm
Nicolas Perrault wrote: Fri Nov 05, 2021 7:14 pm

The P/E 10 (Shiller PE) in 1920 was 4.78 --- the all-time record low (1871-2021 data). This suggested forward returns of 1/4.78 = 20.9% (= earnings yield). Over the next 8.75 years, including dividends, your investment between Dec 1920 and Sep 1929 returned a total real 27.2% annualized (x8.2, using the quotient of Shiller's Total Return Prices for the two dates).

Today the P/E 10 is of 40.00 (source), the second highest it has ever been (after 1999-2000). This suggests forward returns of 1/40 = 2.5%. That's the strength of the price "motor" inside the S&P500 nowadays. Quite different from the price motor in 1920.

For the markets to do x5 in 6-7 years, the P/E 10 would need to reach almost 200. "Given a similar situation..." Let me stop you right there, it's not a similar situation.
1/CAPE was never a thing until about 5-10 years ago, when people noticed that 1/CAPE actually modeled the new line.

For instance, in 1996, a few years after Shiller INVENTED CAPE, when CAPE was 25, he predicted 0% real return, not 4% real.

Because that is what the data said.

The data has changed over the past 30 years, and NOW, 1/CAPE is somewhat close to the NEW line, but don't fool yourself that there is some cosmic law that says 1/CAPE actually has predictive power.
I'm not sure I follow you HomerJ. What "new line" are you talking about? The subsequent annualized 10-year excess return line?
The new line of best fit has changed over the years. 1/CAPE now approximates that line. It didn't in the past. You wrote an entire post using 1/CAPE to calculate forward returns.

1/CAPE didn't fit the data in the past. That's a recent (and likely random) new development.
I have no idea what line of best fit you‘re talking about. Nor have I claimed to have „calculated“ forward returns. No one can do that.

I see from your post history that you have a strong aversion to the CAPE metric because you feel that people have been using it to „predict the past“.You also seem to dislike the argument that „the CAPE is high so it must fall“. Check my post, I did not say this.

What I do say is this. If the CAPE is at 5 today and stays at 5 for the foreseeable future, you are far more likely to get greater returns over the next decade than if it is now at 40 and stays at 40 for the foreseeable future. To produce greater returns in the latter scenario, the earnings growth would have to be massive to make up for the tiny initial earnings yield. You cannot escape this HomerJ, it‘s just math.
But you’re not comparing a cape of 5 today to a cape of 40 today. From your post above, you are comparing a cape of 5 in 1920 to a cape of 40 in 2021. A hundred years have passed, yet you seem to believe that valuations have not.

Also, your above example about a cape of 4.78 in 1920 is really odd and seems to be an extreme case of data cherry picking. For one, I thought the cape was supposed to predict the returns for the following 10 years. Why are you looking at only the next 8.75 years ? This seems very arbitrary. Secondly, the run up in prices in the 20s up to the crash in 1929 that kickstarted the Great Depression was universally viewed as being caused by irrational investors and excessive speculation. Yet based on your post, it seems like you are saying that the increases in the market during this time was totally rational, because the cape was so low in 1920 ? Please correct me if I’m wrong.
User avatar
HomerJ
Posts: 21281
Joined: Fri Jun 06, 2008 12:50 pm

Re: U.S. stocks continue to soar!

Post by HomerJ »

The whole thing has veered off course... Let's just all agree that's it not likely that we will see SP 15,000 - 20,000 in 6-7 years. That's really what Nicolas was responding too in the first place. Someday, sure, but I agree that we're likely not in the same situation as we were in 1920.

I guess I just didn't like the 1/CAPE calculations he was making.
Going back to the roaring twenties, the Dow advanced about 4-5x from the earlier low.

Given a similar situation, I think we could easily see Dow 100,000-120,000, S&P 15,000-20,000 and Nasdaq 50,000 by 2027/2028.
"The best tools available to us are shovels, not scalpels. Don't get carried away." - vanBogle59
nigel_ht
Posts: 4742
Joined: Tue Jan 01, 2019 9:14 am

Re: U.S. stocks continue to soar!

Post by nigel_ht »

Nicolas Perrault wrote: Fri Nov 05, 2021 11:49 pm
HomerJ wrote: Fri Nov 05, 2021 11:07 pm
Nicolas Perrault wrote: Fri Nov 05, 2021 7:46 pm
HomerJ wrote: Fri Nov 05, 2021 7:28 pm
Nicolas Perrault wrote: Fri Nov 05, 2021 7:14 pm

The P/E 10 (Shiller PE) in 1920 was 4.78 --- the all-time record low (1871-2021 data). This suggested forward returns of 1/4.78 = 20.9% (= earnings yield). Over the next 8.75 years, including dividends, your investment between Dec 1920 and Sep 1929 returned a total real 27.2% annualized (x8.2, using the quotient of Shiller's Total Return Prices for the two dates).

Today the P/E 10 is of 40.00 (source), the second highest it has ever been (after 1999-2000). This suggests forward returns of 1/40 = 2.5%. That's the strength of the price "motor" inside the S&P500 nowadays. Quite different from the price motor in 1920.

For the markets to do x5 in 6-7 years, the P/E 10 would need to reach almost 200. "Given a similar situation..." Let me stop you right there, it's not a similar situation.
1/CAPE was never a thing until about 5-10 years ago, when people noticed that 1/CAPE actually modeled the new line.

For instance, in 1996, a few years after Shiller INVENTED CAPE, when CAPE was 25, he predicted 0% real return, not 4% real.

Because that is what the data said.

The data has changed over the past 30 years, and NOW, 1/CAPE is somewhat close to the NEW line, but don't fool yourself that there is some cosmic law that says 1/CAPE actually has predictive power.
I'm not sure I follow you HomerJ. What "new line" are you talking about? The subsequent annualized 10-year excess return line?
The new line of best fit has changed over the years. 1/CAPE now approximates that line. It didn't in the past. You wrote an entire post using 1/CAPE to calculate forward returns.

1/CAPE didn't fit the data in the past. That's a recent (and likely random) new development.
I have no idea what line of best fit you‘re talking about. Nor have I claimed to have „calculated“ forward returns. No one can do that.

I see from your post history that you have a strong aversion to the CAPE metric because you feel that people have been using it to „predict the past“.You also seem to dislike the argument that „the CAPE is high so it must fall“. Check my post, I did not say this.

What I do say is this. If the CAPE is at 5 today and stays at 5 for the foreseeable future, you are far more likely to get greater returns over the next decade than if it is now at 40 and stays at 40 for the foreseeable future. To produce greater returns in the latter scenario, the earnings growth would have to be massive to make up for the tiny initial earnings yield. You cannot escape this HomerJ, it‘s just math.
if earnings is $3 and share prices are $15 and $120 it strikes me that you end up with $30 and $240 if eps goes to $6 when you hold PE constant at 5 and 40.

If you want to argue that it’s more likely for PE 5 to go to 10 than 40 to 80 I’d agree there.
User avatar
Nicolas Perrault
Posts: 244
Joined: Thu Sep 27, 2018 12:33 pm
Location: Oxford, UK

Re: U.S. stocks continue to soar!

Post by Nicolas Perrault »

Soysauceonrice wrote: Sat Nov 06, 2021 12:26 am
Nicolas Perrault wrote: Fri Nov 05, 2021 11:49 pm
HomerJ wrote: Fri Nov 05, 2021 11:07 pm
Nicolas Perrault wrote: Fri Nov 05, 2021 7:46 pm
HomerJ wrote: Fri Nov 05, 2021 7:28 pm

1/CAPE was never a thing until about 5-10 years ago, when people noticed that 1/CAPE actually modeled the new line.

For instance, in 1996, a few years after Shiller INVENTED CAPE, when CAPE was 25, he predicted 0% real return, not 4% real.

Because that is what the data said.

The data has changed over the past 30 years, and NOW, 1/CAPE is somewhat close to the NEW line, but don't fool yourself that there is some cosmic law that says 1/CAPE actually has predictive power.
I'm not sure I follow you HomerJ. What "new line" are you talking about? The subsequent annualized 10-year excess return line?
The new line of best fit has changed over the years. 1/CAPE now approximates that line. It didn't in the past. You wrote an entire post using 1/CAPE to calculate forward returns.

1/CAPE didn't fit the data in the past. That's a recent (and likely random) new development.
I have no idea what line of best fit you‘re talking about. Nor have I claimed to have „calculated“ forward returns. No one can do that.

I see from your post history that you have a strong aversion to the CAPE metric because you feel that people have been using it to „predict the past“.You also seem to dislike the argument that „the CAPE is high so it must fall“. Check my post, I did not say this.

What I do say is this. If the CAPE is at 5 today and stays at 5 for the foreseeable future, you are far more likely to get greater returns over the next decade than if it is now at 40 and stays at 40 for the foreseeable future. To produce greater returns in the latter scenario, the earnings growth would have to be massive to make up for the tiny initial earnings yield. You cannot escape this HomerJ, it‘s just math.
But you’re not comparing a cape of 5 today to a cape of 40 today. From your post above, you are comparing a cape of 5 in 1920 to a cape of 40 in 2021. A hundred years have passed, yet you seem to believe that valuations have not.
I believe you're missing the point.
Soysauceonrice wrote: Sat Nov 06, 2021 12:26 am Also, your above example about a cape of 4.78 in 1920 is really odd and seems to be an extreme case of data cherry picking. For one, I thought the cape was supposed to predict the returns for the following 10 years. Why are you looking at only the next 8.75 years ? This seems very arbitrary. Secondly, the run up in prices in the 20s up to the crash in 1929 that kickstarted the Great Depression was universally viewed as being caused by irrational investors and excessive speculation. Yet based on your post, it seems like you are saying that the increases in the market during this time was totally rational, because the cape was so low in 1920 ? Please correct me if I’m wrong.
I'm looking at 8.75 years because I was responding to someone who claimed that stock returns in the 1920s were enormous. I said "sure they were enormous, but the returns of the 2020s are unlikely to be so high". If I had used 10 years, then the "sure they were enormous" part would not have held for obvious reasons. That's all. I did not claim that the 1920s run-up was rational, only that it was more probable than is a similar future 5X run up in the 2020s due to different starting points.
HomerJ wrote: Sat Nov 06, 2021 1:53 am The whole thing has veered off course... Let's just all agree that's it not likely that we will see SP 15,000 - 20,000 in 6-7 years. That's really what Nicolas was responding too in the first place. Someday, sure, but I agree that we're likely not in the same situation as we were in 1920.
I agree, let's not derail the whole thing :D
HomerJ wrote: Sat Nov 06, 2021 1:53 am I guess I just didn't like the 1/CAPE calculations he was making.
That's fine :sharebeer
User avatar
aj76er
Posts: 1179
Joined: Tue Dec 01, 2015 10:34 pm
Location: Austin, TX

Re: U.S. stocks continue to soar!

Post by aj76er »

HomerJ wrote: Sat Nov 06, 2021 1:53 am The whole thing has veered off course... Let's just all agree that's it not likely that we will see SP 15,000 - 20,000 in 6-7 years. That's really what Nicolas was responding too in the first place. Someday, sure, but I agree that we're likely not in the same situation as we were in 1920.

I guess I just didn't like the 1/CAPE calculations he was making.
Going back to the roaring twenties, the Dow advanced about 4-5x from the earlier low.

Given a similar situation, I think we could easily see Dow 100,000-120,000, S&P 15,000-20,000 and Nasdaq 50,000 by 2027/2028.
S&P 500 P/E = 40 (~2.5% real forward returns)
Total U.S. Market P/E = 35 (~3% real forward returns)
Total World Market P/E = 27 (~4% real forward returns)

Total market indices, while still high, don’t looks nearly as overpriced.
"Buy-and-hold, long-term, all-market-index strategies, implemented at rock-bottom cost, are the surest of all routes to the accumulation of wealth" - John C. Bogle
nigel_ht
Posts: 4742
Joined: Tue Jan 01, 2019 9:14 am

Re: U.S. stocks continue to soar!

Post by nigel_ht »

aj76er wrote: Sat Nov 06, 2021 8:26 am
HomerJ wrote: Sat Nov 06, 2021 1:53 am The whole thing has veered off course... Let's just all agree that's it not likely that we will see SP 15,000 - 20,000 in 6-7 years. That's really what Nicolas was responding too in the first place. Someday, sure, but I agree that we're likely not in the same situation as we were in 1920.

I guess I just didn't like the 1/CAPE calculations he was making.
Going back to the roaring twenties, the Dow advanced about 4-5x from the earlier low.

Given a similar situation, I think we could easily see Dow 100,000-120,000, S&P 15,000-20,000 and Nasdaq 50,000 by 2027/2028.
S&P 500 P/E = 40 (~2.5% real forward returns)
Total U.S. Market P/E = 35 (~3% real forward returns)
Total World Market P/E = 27 (~4% real forward returns)

Total market indices, while still high, don’t looks nearly as overpriced.
Sometimes PE is lower or higher because shares are priced incorrectly. Sometimes it’s lower or higher because the market is right.

The assertion here is that when PE is high the market is usually off its rocker. That may be true but then the adage “nobody knows nothing” is disproven…which would be something of a relief since it’s way overused here.

I keep international and SCV as part of the allocation because while I’m good with US total market as the baseline diversification is good.

40% VTI + 20% VXUS + 10% VIOV strikes me as a better mix than 70% VT. 10% BND + 10% TLT + 5% GLDM + 5% other (HYS, iBonds, TCTSNBN, etc) strikes me as better than 30% BND or TLT.

Is it slightly messier than 3 funds? Yes. Of course I can convert VTI & VXUS to VT if I ever feel the need and there is only seven things anyway and not terribly hard to rebalance.

Seven +/- two is often misused* but I use it anyway.



* The rule is about short term memory only and not necessarily cognitive load beyond one dimensional judgement tasks. But folks apply it to a lot of other stuff because why not? It results in an amount fewer than fingers and generally will fit on one page of whatever (charts, descriptions, etc).
User avatar
LadyGeek
Site Admin
Posts: 95691
Joined: Sat Dec 20, 2008 4:34 pm
Location: Philadelphia
Contact:

Re: U.S. stocks continue to soar!

Post by LadyGeek »

willthrill81 wrote: Fri Nov 05, 2021 8:00 pm I started a thread that became very long in which we discussed this issue. HomerJ posted his views in it often. It might be better to move this discussion into that thread to avoid derailing this one.
Thanks! I moved Nicolas Perrault's discussion from U.S. stocks continue to soar! into this thread.
Wiki To some, the glass is half full. To others, the glass is half empty. To an engineer, it's twice the size it needs to be.
User avatar
cflannagan
Posts: 1208
Joined: Sun Oct 21, 2007 11:44 am
Location: Working Remotely

Re: CAPE: A much stronger predictor of stock returns than many think

Post by cflannagan »

Late to the party here!

Was wondering if by any chance someone in this (12-page-long) thread already shared a graph that shows historical 1/CAPE vs actual 10-year returns after that point? Just so I can visualize how well 1/CAPE does (or doesn't) "fit" the 10 year subsequent returns?

I tried a bit of my Google fu, but I keep coming up with hits on graphs/articles showing relationship between CAPE (not 1/CAPE) and 10 year returns, which isn't really what I was looking for.
User avatar
Topic Author
willthrill81
Posts: 32250
Joined: Thu Jan 26, 2017 2:17 pm
Location: USA
Contact:

Re: CAPE: A much stronger predictor of stock returns than many think

Post by willthrill81 »

cflannagan wrote: Sat Nov 06, 2021 12:14 pm Late to the party here!

Was wondering if by any chance someone in this (12-page-long) thread already shared a graph that shows historical 1/CAPE vs actual 10-year returns after that point? Just so I can visualize how well 1/CAPE does (or doesn't) "fit" the 10 year subsequent returns?

I tried a bit of my Google fu, but I keep coming up with hits on graphs/articles showing relationship between CAPE (not 1/CAPE) and 10 year returns, which isn't really what I was looking for.
In this post in the thread, marcopolo showed how the fitted regression line between CAPE and forward returns has shifted upward since 1989 compared to what it was pre-1989. It's not 1/CAPE, but that's probably an overly simplistic means of using CAPE to estimate forward returns.
The Sensible Steward
000
Posts: 8211
Joined: Thu Jul 23, 2020 12:04 am

Re: CAPE: A much stronger predictor of stock returns than many think

Post by 000 »

The publicly known existence of a model that could accurately predict stock returns (to some degree) would likely commensurately reduce the equity risk premium. As long as stocks remain a risk asset class, there should not exist any perfect model, but valuations probably are on average indicative of the midpoint of the range of possibilities.
User avatar
Topic Author
willthrill81
Posts: 32250
Joined: Thu Jan 26, 2017 2:17 pm
Location: USA
Contact:

Re: CAPE: A much stronger predictor of stock returns than many think

Post by willthrill81 »

000 wrote: Sat Nov 06, 2021 3:13 pm The publicly known existence of a model that could accurately predict stock returns (to some degree) would likely commensurately reduce the equity risk premium.
That could only be true if the market was not efficient.
The Sensible Steward
000
Posts: 8211
Joined: Thu Jul 23, 2020 12:04 am

Re: CAPE: A much stronger predictor of stock returns than many think

Post by 000 »

willthrill81 wrote: Sat Nov 06, 2021 3:38 pm
000 wrote: Sat Nov 06, 2021 3:13 pm The publicly known existence of a model that could accurately predict stock returns (to some degree) would likely commensurately reduce the equity risk premium.
That could only be true if the market was not efficient.
How? If stocks have predictable returns, should they not have less risk?
User avatar
Topic Author
willthrill81
Posts: 32250
Joined: Thu Jan 26, 2017 2:17 pm
Location: USA
Contact:

Re: CAPE: A much stronger predictor of stock returns than many think

Post by willthrill81 »

000 wrote: Sat Nov 06, 2021 3:44 pm
willthrill81 wrote: Sat Nov 06, 2021 3:38 pm
000 wrote: Sat Nov 06, 2021 3:13 pm The publicly known existence of a model that could accurately predict stock returns (to some degree) would likely commensurately reduce the equity risk premium.
That could only be true if the market was not efficient.
How? If stocks have predictable returns, should they not have less risk?
If the market is efficient, it's already aware of any and all models of future returns and has 'priced them' all in to current prices. That's why the similar claim that acknowledging the existence of a factor will erode that factor's future returns also assume some degree of market inefficiency because the market should have known about the factor all along.
The Sensible Steward
Post Reply